What is Going On with Sampling in EU? Challenging the MDR's Stance on Technical Documentation Review

The European Union's Medical Device Regulation (MDR) and In Vitro Diagnostic Regulation (IVDR) introduced a rigorous system for conformity assessment, but the policy of Technical Documentation (TD) sampling has become a lightning rod for debate. While intended as a pragmatic way for Notified Bodies (NBs) to manage the enormous volume and variety of devices, this policy, primarily guided by MDCG 2019-13, risks undermining the regulation's core promise of comprehensive device scrutiny. This article challenges the current mindset and urges manufacturers to adopt a global, diverse compliance strategy.

1. Background: The How and Who of EU Technical Documentation Sampling

Under the MDR and IVDR, the responsibility for assessing compliance for most devices (Class IIa, IIb, III MDR and Class B, C, D IVDR) falls to Notified Bodies (NBs). Given the vast number of devices a manufacturer might produce, the NB implements a sampling plan for the TD review.

This process is governed by a set of criteria:

Initial Certification: The NB must assess the TD of at least one representative device (Basic UDI-DI) per category of devices (for Class IIa/B) or generic device group (for Class IIb/C). A generic device group is defined by devices sharing the "same technical concept and serve similar intended purposes".
Surveillance: Over the five-year validity period of the certificate, the NB is expected to sample and assess 15% of the devices from each group/category. This number can be reduced to a minimum of 5% in the first certification cycle or until the MDCG guidance is updated. A minimum of one TD must be reviewed annually.
Qualitative Factors: The sampling plan must also consider factors like the novelty of the technology, similarities in design, and the outcomes of post-market surveillance (PMS) and vigilance data reviews.

2. The Flaw in the System: Why Sampling Is a Gap in the MDR Mindset

The negative idea that must be brought to the forefront is this: the current TD sampling mechanism is fundamentally ill-suited to the vast technical and clinical diversity of modern medical devices, even within a single regulatory group.

The core flaw lies in the assumption of true representativeness. A generic device group is supposed to share a "same technical concept" and "similar intended purposes". However, in reality, a single device class or group can encompass products with fully different clinical focuses and technical risks.

For example, a "generic device group" of surgical instruments (Class IIa) could include one TD for an arthroscopic tool and another for a neurosurgical instrument. While both are "surgical instruments," the required TD focus—covering materials, biocompatibility, sterilization, and most importantly, clinical evidence (CER)—will be entirely distinct. Sampling one of these TDs risks missing critical non-compliance in the unsampled device due to a completely different intended use scenario.

The MDR's architects and Notified Bodies have implicitly acknowledged the massive product volume and are attempting to balance "thorough regulatory scrutiny and operational efficiency", which often translates to the unspoken excuse of limited capacity. However, if the EU regulation's ultimate goal is enhanced patient safety, the risk of a "gap" where devices with diverse intended uses are simply missed by a 15% quota is too high. This is a crucial gap in the MDR: a policy designed for logistical efficiency that compromises comprehensive safety review across highly diverse product ranges.

3. Manufacturer's Action: Moving Beyond a Sampling Mindset

For manufacturers, the sampling policy should not be seen as a quota to satisfy, but as an alarm bell. To achieve a higher, "MDR level" of compliance across all products and adopt a needed international mindset of diverse compliance, manufacturers must prepare every single TD as if it will be the one selected for a full audit.

Key actions to go beyond the sampling mentality include:

Rigorously Consistent TD Architecture: Ensure all TDs, even those in the same group, are built on a robust, standardized framework
Traceability for Every Device: Prepare a comprehensive General Safety and Performance Requirements (GSPR) checklist for each device/Basic UDI-DI, providing clear, traceable links (document name, version, page number) to the evidence for every single requirement, even if similar to another device in the group.
Develop a Global Compliance TD: The most strategic move is to prepare a single Technical File/Design Dossier that concurrently satisfies the most stringent global requirements (e.g., MDR and FDA requirements), ensuring that all technical, clinical, and regulatory differences are addressed in a unified document from the design phase.

4. Notified Body's Role: Checking the Unsampled Device

The NB's check on the unsampled majority of devices occurs primarily through the Quality Management System (QMS) audit and Post-Market Surveillance (PMS) review. The sampling itself is arguably an audit of the QMS's output—a verification that the TD generation process is robust—rather than a full product safety review for every device.

To verify compliance for devices not sampled, NBs must:

Deepen the QMS Audit: Focus not just on the QMS procedures but on how the manufacturer manages the diversity of their product range within the QMS.
Scrutinize Risk Management and PMS Data: The NB must assess the Post-Market Surveillance (PMS) Plan, Periodic Safety Update Reports (PSURs), and vigilance data for the entire device range annually. Any signal of a problem in an unsampled device should trigger an immediate re-sampling and in-depth review of that device's specific TD.

By focusing on the outputs of the QMS and the feedback from the PMS loop for all devices, the NB can close the gap created by the quantitative sampling policy.

5. The Urgency: An International Mindset of Diverse Compliance

The MDCG 2019-13 sampling guidance is a pragmatic regulatory measure born of necessity. However, it harbors a dangerous gap in the system where genuine diversity in intended use and clinical application is masked by an administrative grouping (generic device group). The idea that sampling is needed because "same content among TD must be avoided" is flawed; while duplication must be avoided, a unique risk-benefit profile and clinical evaluation for every distinct product configuration is non-negotiable for patient safety.

The solution is not to wait for the MDR to demand 100% TD review, which is logistically impossible. Instead, manufacturers must adopt an international mind of diverse compliance, preparing every TD with the rigor of a Class III/PMA submission, ensuring that their internal processes—the QMS—guarantee safety for all their diverse products, sampled or not. This is the only way to genuinely meet the spirit of the MDR and safeguard global market access in a post-sampling regulatory world.

Every compliance has its history, as MDR sampling in Europe. As in some markets it lacks system of product code, product specific guidance and standard in EU.

We hope that every EU-manufacturer maintain all technical documentation with global overview.

For non EU-manufacturing obtaining CE it is first to choose a typical TD to send it to notified body to prove.

If one day EU law-maker would like to find a way to cover the gap the TD-sampling, we are keen to sharing our global experience.